Our long-term goal is to understand the emergence of pandemic influenza viruses, such as the 1918 influenza virus that caused the most devastating outbreak of an infectious disease on record. In particular, we are asking two questions: (i) how likely is the emergence of a pandemic strain comparable to the 1918 virus?; and (ii) what are the molecular features that accounted for the unprecedented virulence of the 1918 virus? Our studies have revealed that avian influenza virus genes that encode 1918 virus-like proteins still exist in nature, suggesting that 1918-like pandemic viruses may emerge in the future. In Aim 1, we will therefore assess the risk for the emergence of 1918 virus-like pandemic influenza viruses. To assess this risk, we have generated a 1918-like avian influenza virus comprising avian influenza viral segments that encode 1918-like proteins. We will introduce 1918 viral amino acids into this virus to identify changes that are critical for pathogenicity. In a parallel approach, we have generated mutant 1918 viruses in which one gene at a time is replaced with its closest avian relative (as determined by protein homology). The introduction of 1918 amino acids into the 'avian' gene/protein will reveal amino acids that are critical for 1918 virus pathogenicity. We and others have identified the HA gene as an important determinant of the unprecedented virulence of the 1918 influenza virus. The HA protein carries out two critical functions in the viral life cycle - receptor-binding and fusion of the viral and endosomal membranes. Currently, it is not known if the 1918 HA protein differs in its receptor-binding properties and/or fusion kinetics from contemporary human viruses, and if potential differences contributed to the high virulence of the 1918 virus. In Aim 2, we will therefore test whether the 1918 HA protein differs from that of contemporary human viruses in its receptor-binding properties. These features may be affected by the HA glycosylation pattern, prompting us to test the significance of the atypical 1918 HA glycosylation pattern for virulence. In addition, we will characterize the fusion kinetics of the 1918 HA protein, with particular emphasis on the characterization of a previously unrecognized histidine patch in the 1918 HA protein that is thought to affect fusion. In addition to the HA gene, the viral replication complex has been identified as a determinant of host range and pathogenicity. Our preliminary data also suggest a role for the 1918 virus replication complex in the virulence of the 1918 virus. However, the exact contributions of the components of the replication complex to increased virulence and the underlying mechanisms (including effects on host cell responses) are currently unknown, a gap in knowledge that we propose to fill in Aim 3. Collectively, these studies are designed to address two critical questions in influenza virus research - why was the 1918 virus highly pathogenic, and what is the risk for emergence of another 1918-like virus?